Political Cross-Market Arbitrage (Arbiter)
Hunts no-arbitrage violations between related Polymarket election markets, sizes positions with fractional Kelly, and only trades with backtested out-of-sample edge.
This quant bot scans for arbitrage opportunities across related political prediction markets on Polymarket. For example, if market A says 'Candidate X wins state Y' at 60¢ and market B says 'Candidate X wins presidency' at 55¢, but the joint probability violates no-arbitrage bounds (e.g., P(win presidency) must be at least P(win state Y) * some factor), the bot places paired trades to lock in a risk-free profit. It uses fractional Kelly criterion to size positions, balancing growth with risk of ruin. Crucially, it refuses to trade unless the strategy has been backtested on out-of-sample historical data showing a statistically significant edge. The bot paper-trades by default, prioritizing survival over profit. It uses a process-first approach: position sizing logic is rigorously tested before any live deployment.